'AuraRing': Electromagnetic Finger Tracking Wearable

Researchers at the University of Washington  have developed the "AuraRing" ring and wristband wearable device that detects the exact location of a user's index finger and continuously tracks hand movements. AuraRing could detect the onset of Parkinson’s disease by tracking subtle hand tremors, or help with stroke rehabilitation by providing feedback on hand movement exercises. The ring emits a signal that is picked up by the wristband, which can then identify the position and orientation of the ring (and the finger). AuraRing is made up of a coil of wire wrapped 800 times around a 3D-printed ring. A current running through the wire generates a magnetic field, which is detected by three sensors on the wristband. The device can even pick up handwriting, potentially for quick responses to text messages, and could enable the user to have a virtual reality avatar hand.

Transcript

00:00:05 we present ordering a wearable input device that enables high fidelity finger tracking our ring uses custom electromagnetic tracking to precisely and continuously estimate the five degree of freedom pose of a ring from a wristband this enables freeform and subtle input for wearable platforms like smartwatches or mixed reality headsets our ring relies on a short range and low

00:00:26 power AC electromagnetic tracking technique the system consists of a wristband which contains three sensor coils and a ring which acts as a magnetic field generator to minimize the size of the ring the magnetic field is generated using a thin wire coil that is tightly wrapped around the surface of the ring the ring generates an oscillating magnetic field around the

00:00:47 user's hand which is measured by three rigid sensor coils in the wristband as the user moves their finger the voltage induced in each of the sensor coils changes resulting in the measurements that Oren uses to reconstruct the pose of the ring in order to perform this accurately our ring uses a one-time calibration procedure to empirically learn parameters of the sensor model

00:01:05 which describes how a given pose manifests as sensor measurements pose reconstruction and runtime is then accomplished using either a nonlinear optimizer or a closed form neural network or ring focuses exclusively on short ring tracking in the range of 10 to 15 centimeters it has a sensor resolution of a tenth of a millimeter and a typical accuracy of four point

00:01:26 four millimeters this enables applications like freeform drawing or handwriting thanks to a high bandwidth and data rate the system is also able to detect high-speed events like taps of various intensities in this example users tap on the left and right very quickly and the or ring can give immediate feedback due to their broadband spectrum and higher energy by

00:01:45 leveraging a kinematic model of the hand or a ring tracking can be used to drive an inverse kinematics model to estimate hand pose this can enable interaction with virtual objects for mixed reality headsets by combining aura ring with an arm or wrist tracking solution items can be selected in absolute coordinates we envision or ring as a way to extend the interaction modalities for many use

00:02:07 cases oaring enables precise continuous finger tracking which can be used to better interact with smartwatches which have otherwise limited input areas as a magnetic tracking device our ring can be particularly useful for subtle input as no line of sight is required in augmented and virtual reality applications Oran can improve immersiveness by allowing for

00:02:27 the detection of interactions such as taps swipes and subtle gestures a user can select and manipulate virtual objects with their finger